regexp to leave only alpha/numeric chars

Using preg_replace() is there a simple regexp to strip everything from a
string except alpha and numeric chars (a-zA-Z0-9)?

$input = "$tring1!";
$pattern =
$input = preg_replace($pattern, "", $input);

result: "tring1"

Bosconian wrote:

> Using preg_replace() is there a simple regexp to strip everything from a
> string except alpha and numeric chars (a-zA-Z0-9)?
>
> $input = "$tring1!";
> $pattern =
> $input = preg_replace($pattern, "", $input);
>
> result: "tring1"

Chances are anything you will ever try to do with regexp's has been done,
and there may even be a shortcut for it. So at this page:

http://www.php.net/manual/en/referen...ern.syntax.php

you would be able to find this:

$input = preg_replace('\W','',$input)

This will leave in underscores, but that is easily fixed:

$input = preg_replace('[\W_]','',$input)

Also try reading an introduction to regexp's in a Perl guide, you can likely
find something more of a tutorial there to get you started.

--
Kenneth Downs
Secure Data Software, Inc.
(Ken)nneth@(Sec)ure(Dat)a(.com)

"Kenneth Downs" <knode.wants.this@see.sigblock> wrote in message
news:3ifrj2-q7n.ln1@pluto.downsfam.net...
> Bosconian wrote:
>
> > Using preg_replace() is there a simple regexp to strip everything from a
> > string except alpha and numeric chars (a-zA-Z0-9)?
> >
> > $input = "$tring1!";
> > $pattern =
> > $input = preg_replace($pattern, "", $input);
> >
> > result: "tring1"
>
> Chances are anything you will ever try to do with regexp's has been done,
> and there may even be a shortcut for it. So at this page:
>
> http://www.php.net/manual/en/referen...ern.syntax.php
>
> you would be able to find this:
>
> $input = preg_replace('\W','',$input)
>
> This will leave in underscores, but that is easily fixed:
>
> $input = preg_replace('[\W_]','',$input)
>
> Also try reading an introduction to regexp's in a Perl guide, you can
likely
> find something more of a tutorial there to get you started.
>
> --
> Kenneth Downs
> Secure Data Software, Inc.
> (Ken)nneth@(Sec)ure(Dat)a(.com)

Ken,

I was able to use your example, but not without adding starting and ending
delimiters:

'/[\W_]/'

Nothing was stripped without them.

I will bone-up on my regexp syntax.

Many thanks.

Bosconian wrote:
> I was able to use your example, but not without adding starting and ending
> delimiters: '/[\W_]/'

You might find '/[\W_]+/' faster. Also, on the linked doc page it says
the \w and \W are locale specific; there might be accented characters
left behind. If you don't want that, use '/[^a-z0-9]+/i'.


--
Firefox Web Browser - Rediscover the web - http://getffox.com/
Thunderbird E-mail and Newsgroups - http://gettbird.com/

Bosconian wrote:

> "Kenneth Downs" <knode.wants.this@see.sigblock> wrote in message
> news:3ifrj2-q7n.ln1@pluto.downsfam.net...
>> Bosconian wrote:
>>
>> > Using preg_replace() is there a simple regexp to strip everything from
>> > a string except alpha and numeric chars (a-zA-Z0-9)?
>> >
>> > $input = "$tring1!";
>> > $pattern =
>> > $input = preg_replace($pattern, "", $input);
>> >
>> > result: "tring1"
>>
>> Chances are anything you will ever try to do with regexp's has been done,
>> and there may even be a shortcut for it. So at this page:
>>
>> http://www.php.net/manual/en/referen...ern.syntax.php
>>
>> you would be able to find this:
>>
>> $input = preg_replace('\W','',$input)
>>
>> This will leave in underscores, but that is easily fixed:
>>
>> $input = preg_replace('[\W_]','',$input)
>>
>> Also try reading an introduction to regexp's in a Perl guide, you can
> likely
>> find something more of a tutorial there to get you started.
>>
>> --
>> Kenneth Downs
>> Secure Data Software, Inc.
>> (Ken)nneth@(Sec)ure(Dat)a(.com)
>
> Ken,
>
> I was able to use your example, but not without adding starting and ending
> delimiters:
>
> '/[\W_]/'
>
> Nothing was stripped without them.
>
> I will bone-up on my regexp syntax.
>
> Many thanks.

I would love to say I left the delimiters out on purpose to improve your
skills, but it would be a lie. I just forgot.

Glad it helped.


--
Kenneth Downs
Secure Data Software, Inc.
(Ken)nneth@(Sec)ure(Dat)a(.com)

"Ewoud Dronkert" <firstname@lastname.net.invalid> wrote in message
news:426b5a6e$0$157$e4fe514c@dreader4.news.xs4all. nl...
> Bosconian wrote:
> > I was able to use your example, but not without adding starting and
ending
> > delimiters: '/[\W_]/'
>
> You might find '/[\W_]+/' faster. Also, on the linked doc page it says
> the \w and \W are locale specific; there might be accented characters
> left behind. If you don't want that, use '/[^a-z0-9]+/i'.
>
>
> --
> Firefox Web Browser - Rediscover the web - http://getffox.com/
> Thunderbird E-mail and Newsgroups - http://gettbird.com/

Even better. Thanks for the tip.